FTP and Elevation: Relevant to Lake Tahoe (and Boulder)
I haven't seen an "official" discussion of this started...maybe I missed it... but in case not, I thought I would begin one. I'm not going to be racing at Tahoe, but I know a bunch of us are. We need to know what to do. Everybody knows that your effort will be affected by altitude, but not how much. Here's the thing I worry about: If your high-elevation FTP is lower than your training-elevation FTP, you may bike at too high a power. Having done that, you've put yourself behind the eightball already, i.e., now you have to run a marathon at elevation, having worked too hard on the bike. Sounds ugly. Also sounds like something a lot of people are going to do.
So the question becomes, "What kind of correction should I expect to have to make?"
I do not claim to be an expert on this, but I did a bit of poking around. This started when I heard an interview with Joe Friel on IM Talk recently. Whatever else we all think about Joe Friel, at least he does speak power language and he self-monitors. What he basically said was that when he moves from Scottsdale (elevation 1500 ft) to Boulder (elevation 5400 ft), his FTP initially drops about 8%, and over the course of a few weeks, he acclimatizes, but the number is never as high as at low elevation. He maximizes at about 4% below his low elevation FTP.
This coincides what I have read over the years about elevation training, i.e., that actual elevation training doesn't do you nearly as much good as you would think because you can't ever go as hard at elevation as you can at low elevation. (Theoretically, if you live high and train low, you can get a benefit, but no real people can do this.)
On the web, I found reference to elevation on twowheelblogs.com here: http://www.twowheelblogs.com/how-altitude-affects-power-output
What the (anonymous?) author of that entry says they did was look around at several studies on FTP vs altitude and "average" the values he found. Below is a reproduction of the first several entries of his table, which represents UN-acclimatized FTP
| Feet | Meters | %FTP |
| 0 | 0 | 100 |
| 1000 | 305 | 99 |
| 2000 | 610 | 98 |
| 3000 | 915 | 96 |
| 4000 | 1220 | 95 |
| 5000 | 1525 | 93 |
| 6000 | 1830 | 92 |
| 7000 | 2135 | 90 |
| 8000 | 2440 | 88 |
| 9000 | 2745 | 86 |
This table is at least in qualitative agreement with Friel's anecdotal quote of an 8% drop in Boulder. Tahoe is at 7200 ft. Depending on where you live, this could mean an 8-10% drop in your effective FTP.
Friel has an older blog post here: http://www.joefrielsblog.com/2010/09/altitude-and-aerobic-performance.html I think there is at least on error in it, because he suggests you can get a benefit if you train high and race low...which I am pretty sure (as noted above) is now thought absolutely NOT to be the case. Nonetheless, he does cite a couple of original papers:
1. Bassett, D.R. Jr., C.R. Kyle, L. Passfield, J.P. Broker, and E.R. Burke. Comparing cycling world hour records, 1967-1996: modeling with empirical data. Medicine and Science in Sports and Exercise 31:1665-76, 1999.
2. Peronnet, F., G. Thibault, and D.L. Cousineau. A theoretical analysis of the effect of altitude on running performance. Journal of Applied Physiology 70(1): 399-404, 1991.
He extracts a similar table as above. For 7000 ft, he cites 86.5% "Available aerobic power" from Basset, and 91.6% from Peronnet (both unacclimatized).
What I hope this thread becomes is something where others who have thought about this and/or done some better research than me can contribute and we can turn it into a wiki entry or at least a part of the Race-Specific info for Tahoe.
Good luck to all my high elevation friends!
William
Comments
Thanks for starting this, William.
Since 2005, I've spent 2-6 weeks a year training for IMs at 5700-9600' while living @ 8400' outside Aspen, CO. Also, I've reviewed some literature about training and racing at altitude. Here are a few notes based on that.
Let's start by noting that the elevation of the lake itself is 6200'. While the bike course does top out over 1000' above that, most of the bike and all of the run is done at or near lake level. So 2000 meters is what we're talking about (best to use metric as all studies are done on that basis.) At that altitude, atmospheric pressure is 20% less than at sea level. Since oxygen crosses into the blood stream across lung membranes mostly by being "pushed" by that pressure, and in part by being "pulled" by the lower concentration of oxygen in venous red blood cells compared to arterial, the effective oxygen in the blood stream at 2000 m is about 10-15% less than at sea level. That variance will depend on how much you are acclimated with increased red cell numbers, size, and blood volume, as well as some other physiologic changes which occur with living and/or training at altitude.
One article I checked out (from "Aviation, Space, and Environmental Medicine", Vol 69, #8) reviewed a number of studies of athletic performance at altitude. Here's a relevant quote: "...at 2000 m altitude, events lasting less than 5 min ... would be minimally affected while events lasting 20-30 min ... would be impaired by about 5%, and events lasting 2-3 h would be impaired by 10-15%." There are a number of factors at work here, but most of us are probably just interested in the actual impact on performance and not the physiology behind it.
All that I have read, and my own personal experience, are consistent with the table Wm includes in his post. Eg, I have a standard TT I do up Maroon Creek Road, going from 8000'-9300' over 6.7 miles at a constant uphill gradient maxing @ 7-8%. Takes me about 40-45 minutes. While I can't do that at 100% of my sea level FTP, I con routinely do it at 85-90% of my FTP.
As to running, I checked back in my log and compared long runs within two weeks of each other of similar distance (within 1-2 miles same distance, greater than 13 miles), done at altitude and at sea level. All were 10-30" (30" >> 6%) slower at an altitude of about 6500' compared to 200' I think the truest difference is more like 30", considering I do most of my long runs on a hilly course at home, and on a flat bike path at altitude. The most recent example bears this out. During my current marathon hack, I did two 18 mile runs two weeks apart, both on bike paths. At sea level, I averaged 8:16; at altitude, I averaged 8:48. So, again, at 2200 meters I seem to run about 5-6% slower over the same distance at similar effort levels at terrain, altitude compared to sea level. YMMV, as I am probably a very fit runner. Others newer to running or slower may have a greater decrement; someone younger and/or fitter may have less. But not a lot less.
My recommendation for Tahoe: use an FTP of 8-10% lower for biking (remember, you will get a slight advantage from less air resistence), and a run pace of 5-10% slower, depending on how strong a runner you are. Altitude is like heat: you can't overcome it with will power alone. Like running up a hill or into the wind on a bike, you just have to back off as much as the environment demands.
Swimming is a separate topic. There's not much data on this, and I don't have long OWS experience of my own to compare. But I do know the biggest risk is starting out at too high of an effort level. You will probably NEVER recover from that during the swim.
Another effect of altitude to remember: the lower air pressure means dryer air. Which means faster insensible loss of water from the lungs, as well as sneaky sweating, meaning it evaporates before you realise your are sweating. So you need more fluid than you normally would for a given effort level at a given temperature. Also, part of acclimitization which you'll experience immediately upon arrival is weight loss which is really dehydration. You'll need to drink more than normal during your first 3-4 days after arriving at altitude.
Finally, less air means more sun radiation gets through. So sun screen is critical. And less air means less heat is retained when the sun is not shining. So a cloudy day will feel colder, going from sun to clouds will chill one, and when the sun goes down, it gets colder quicker.
Training at altitude is another topic entirely, which I can comment on if anyone is interested. Teaser: I believe there IS a benefit to training high, for low altitude events at least for LOOONG events like marathons and Ironman, but you have to know how to do it. For shorter runs and tris, I think it IS true that training at altitude is not a great idea. And, of course, training @ 6200' for an event @ 6200' makes some sense, again, if it is a long event. Not a good idea for sprinters or swimmers, though.
The swim is an interesting point. In a sense, it may be an advantage that you don't get any mid-stream data so you won't feel bad about going a bit slow. What are your plans, Al? I have to think that RPE, focusing on the aerobic feel (and HR, as much as one can "feel" that), would be the way to start for us lowlanders...but you have obviously put a lot more thought into it than I have. I would think those who focus on the RPE of their "drive system" (lats, arms, shoulders) are going to end up overdoing it.
Trivia that shouldn't affect your approach to the race:
Regarding wind resistance on the bike, I went to the analytical cycling calculator site (http://www.analyticcycling.com/ForcesSpeed_Page.html) and fed it a bit of data. Comparing an air density of 1.00 kg/m^3 (guestimate for 2000 m) to the sea level density of 1.22 kg/m^3, you get a speed increase of about 6.7% on the flat and about 2.8% on a 3% incline. It's obviously less on the incline because a greater proportion of the resistance is gravity. (I used 200 W as input; you'd get different numbers if you used grossly different values.)
WJ and Al -
Thanks for the excellent information! My notes:
Finally, 6200-7200 isn't all that bad, altitude-wise, for very, very fit IM athletes (all of you). In my experience, things definitely change, and change very quickly, north of 8k. That is, the difference between 8k and 10k is about 2x the difference (whatever that means) between 6k and 8k.
In summary:
Re: swimming...I have had experience racing in Lake Tahoe, and at higher altitudes in Xterra events. Swimming in a race is ALL RPE anyway, and the only note is to not get into oxygen debt at the start. Frankly, I'm more worried about wind/chop than altitude. One Xterra there, I went about 28 min/1500 m, the other, over 32, at a time when I was doing 26/7 at sea level. Only difference was a big wind out of the South, with massive chop.
Re:air resistence on bike. Less air resistence does not help the FTP issue. It may mean we wont go 10% slower even though we're only able to generate 10% less power on the same RPE. All the more reason to ignore speed, and focus on watts during the bike. Let the speed/time sort itself out.
One other caution...the decreased O2 means the cost of working just a little too hard up those 9-13% grades is greater. Anyone getting close to or above altitude adjusted FTP for more than 2-3 minutes will find recovery happens more slowly, or maybe not at all, just like in swimming. Just one more chnace to not be stupid, as Rich would say.
As a recent new resident to Boulder CO (coming from NJ) I can personally attest to all the info in this thread. FTP has been down, running has been slower and it has taken me months to get my times better.
But the biggest struggle has been with swimming - Al is absolutely correct that if you go too hard at the start - it's almost impossible to recover. I would highly advise all those racing at altitude to just start off slow and build with what you can. Don't expect the same swim times - if you are not used to altitude, it will be really hard to recover. The only time I've had to roll on my back and take a break swimming in a race was in the Steamboat CO Triathlon (at 7000 ft) when after 400 yards I go a little dizzy and was way out of breath. Almost passed out at T1 and never did recover. Don't make this mistake - a few extra minutes in the swim won't hurt....
Once again this team proves to be such a valuable source of information!
A few things that are noticeable during exercise at altitude is you feel it in the lungs before you feel it in the legs. That gives you the perception that you are working harder than you should. Even on the slower runs the lungs seem to run interference. It just burns much more in the lungs. Now, when I run in NC, I feel it in my legs and I don't breathe so crazy out of control, it feels quite easy in comparison. I just don't get that same feeling here. It throws my RPE way off and I would venture to say my RPE is still off. When I run on feel, it just doesn't feel as hard. All my z4 runs I run on feel, I keep just under "negative talk" zone, which is way too fast for the interval. I'm not sure when that will catch up.
My experience with swimming was exactly how Coach, Al, WJ et al described it. If you go out too hard you will be in for some hurt. I raced (Oly) Show Low three times. On the times I went out too fast, I just couldn't recover. I couldn't recover in transition or initially on the bike. It actually took awhile for my legs to feel normal riding. My legs would have a tingling feeling. In contrast, when I went out easy "ish" and built into the swim. I ended up finishing strong and it led to an easier transition. The bike didn't have the crazy legs and ended up feeling better. Starting out easy will be important.
On the bike, I don't think we will lose as much time as if we solely reduced our power compared to sea level. The thinner air will allow us to move quicker at the same level of watts. I know we are planning on reducing watts but there is a measurable difference of speed in the thinner air. I believe Moriarty, NM held the 1 hour record for awhile and it still draws people nationally to race it. Thinner air is faster. Reducing the wattage will slow us down but by how much, not sure. When I have used bike calculator and I input a ride that I just completed but then change the altitude from 600 ft to 6000 ft, with those same parameters, it says a 1.7 mph increase. I'm not buying THAT but based on all my riding and racing, 200 watts at sea level vs 200 watts at altitude. Altitude will be faster. All that being said, I am also going by what Al said, follow my intensity guide (power) and just watch that. Speed will be what it is.
The run I am a little unsure about. Which to me is both good and bad. Normally for a given pace, altitude is a 10 beat per minute increase. But, running here in the heat and humidity leads to some dehydration especially on the long runs. This (exercising in the head and humidity with dehydration) has been proven to give an increase in performance (in colder conditions) by increased blood plasma volume. One of the big performance limiters for me is the heat and humidity. With the higher dew points it is harder to cool my body thus it works that much harder to cool it. Going to altitude will have a much lower dew point and way cooler, but my body will be working harder to get the oxygen it's used to. Some research I read and watched was it generally takes several weeks to counter the affects of altitude and it never reaches a level of sea level output. There was a thought that you could fly in and fly out. Meaning show up the day of and just race before the negative effects of altitude hit you. The test concluded there was a 6 hr window before you were hindered. The problem is that as soon as you hit altitude the negative aspects start. At 6 hrs, however, you are at the worse than any other time in the physiological markers they tested. It does get better through acclimation. I basically went through the you tube videos of the altitude training and team sports conference that occurred this year and a pub med document.
There were some other bits of information that may change my race morning tactics. But I can bring that up in another post. I have rambled on too long.
For those wanting citations.
http://www.ncbi.nlm.nih.gov/pubmed?Db=pubmed&Cmd=ShowDetailView&TermToSearch=11224821&ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum
http://www.ncbi.nlm.nih.gov/pubmed?Db=pubmed&Cmd=ShowDetailView&TermToSearch=16874590&ordinalpos=2&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum
Some (likewise personal) observations after reading Steve's post: